Real Search Console and rank tracker screenshots, with the date range and method for each. Tap a number to see the screenshot it comes from.
Merged duplicate collection paths and stopped filter pages from being indexed for an international store. Clicks went from 2.74K to 18.4K, 3 months vs the 3 before.
Merged regional service pages that competed for the same searches into one page per buyer intent. Clicks went from 5.7K to 23.3K, 3 months vs the 3 before.
Made the business details consistent across directories and rebuilt the Business Profile. After the work, 71 of 81 grid points ranked in the top 3.
Moved the menu out of a PDF into one page per category, fixed speed and mobile, and cleaned up the Business Profile. Clicks went from 5.6K to 11.2K, 6 months vs the 6 before.
Wrote pages for narrow developer problems and tool comparisons and served them as plain HTML. The new site earned 610 clicks from 15 May to 26 July 2026.
Key takeaway
These SEO case studies each rest on one real screenshot from Google Search Console or a map rank tracker. Every number links to that screenshot, every case explains what was done, and every case says what it cannot prove. If a result has no screenshot, it is not on this page.
Most SEO case studies online show a rising graph and a big percentage, and stop there. That is not enough to judge whether the work caused the change, or whether it would work for you.
So every case here follows the same rules:
When you tap a number on any case page, you see the screenshot it comes from. When a figure cannot be traced to a screenshot, I leave it out, even when it would make the case look better.
Six cases have a full page so far. Each summary below gives the problem, the work, the one number that matters, and the main limit. The full page for each case shows the screenshot, the table of figures, and the method in more detail.
An international store: structure before content. A specialty retailer had the same collections at several addresses, filter combinations that Google could index, and catalog pages with no self referencing canonical. Fixing those three things came first. Comparing two 3 month periods in Google Search Console, organic clicks grew . The screenshot shows no exact dates, and the store sells internationally, not in Pakistan. Read the full ecommerce structure case.
A B2B services firm: merging pages that competed. A firm serving North America and the UK had several regional versions of the same service page, all chasing the same searches. We merged them into one page per buyer intent, rebuilt the business schema, and worked on speed. Organic clicks grew across two 3 month periods. The screenshot measures clicks, not leads. Read how the competing service pages were merged.
An electrician in New Orleans: the map, measured properly. The business rarely appeared in the top three map results. We made its details consistent across directories, rebuilt the Business Profile, and added local pages with schema. Across a 9 by 9 geo-grid, its average Map Pack rank moved from . The two grids are centred on different parts of the city, so the comparison is indicative, not like for like. Read the geo-grid case with both scans.
A large information site: reach after a clean up. A media site had a big library buried under thin tag and pagination archives. We pruned and merged the archives, rebuilt topic hubs, and reconnected older articles. In one 3 month window the site recorded impressions. The screenshot shows one period only, so it supports reach, not growth. Read the content site case and its limits.
A restaurant: a menu Google could read. The menu sat in a PDF or on one long page, and the site was slow on phones. Speed and mobile came first, then a page for each menu category, pages for local food searches, and a cleaner Business Profile. Comparing two 6 month periods, organic clicks grew . The screenshot measures clicks, not orders or bookings. Read the restaurant menu pages case.
A new developer tool: early traction on a fresh domain. A young SaaS product had no links and no search history. We wrote pages for narrow developer problems and tool comparisons, linked guides to the product, and served every page as plain HTML. From 15 May to 26 July 2026 the site earned clicks. The screenshot shows one window, not growth, and no signup figures. Read how a new SaaS site found its first searchers.
These five screenshots come from other client projects. They are real Search Console reports, but none has a full case page yet. For three of them the industry and the method are not published, so they show a result without explaining it. Each table gives the figures exactly as the screenshot shows them, the period, and the main limit.
| Metric | Previous 3 months | Last 3 months | Change |
|---|---|---|---|
| Clicks | 9.62K | 18.1K | +88% |
| Impressions | 1.31M | 2.16M | 1.6 times |
| Average CTR | 0.7% | 0.8% | +0.1 points |
| Average position | 16 | 9.5 | 6.5 places higher |
Organic clicks grew , comparing the last 3 months with the 3 months before. The screenshot shows no exact dates. The industry and method for this project are not published, and the figures measure clicks, not leads.
| Metric | Previous 6 months | Last 6 months | Change |
|---|---|---|---|
| Clicks | 558 | 5.42K | 9.7 times |
| Impressions | 26.5K | 479K | about 18 times |
| Average CTR | 2.1% | 1.1% | -1.0 point |
| Average position | 45.5 | 11 | 34.5 places higher |
Organic clicks grew , comparing the last 6 months with the 6 months before. Average CTR fell because impressions grew much faster than clicks. The screenshot shows no exact dates, and the industry and method are not published.
| Metric | Value |
|---|---|
| Clicks | 3.56K |
| Impressions | 155K |
| Average CTR | 2.3% |
| Average position | 8.6 |
The site recorded clicks in one period that starts on 30 May 2026 and runs into early October 2026. The daily line rises across the chart, but there is no comparison period, so this shows reach in one window, not growth. The industry and method are not published.
| Metric | Value |
|---|---|
| Clicks | 2.05K |
| Impressions | 204K |
| Average CTR | 1% |
| Average position | 11.5 |
In a 3 month window starting 10 April 2026, the site recorded impressions. The work, from my notes on a project shown on my earlier website, fixed thin and duplicated location pages, linked the Business Profile with the city pages, rebuilt the location pages, and added local business structured data. The screenshot shows one window, not growth, and it does not show the work.
| Metric | Value |
|---|---|
| Clicks | 1.31K |
| Impressions | 985K |
| Average CTR | 0.1% |
| Average position | 56.8 |
In one period the directory recorded impressions. The work, from my notes on the same earlier website, moved category listings to server rendered HTML and pointed filter parameter URLs to clean category pages with canonicals. The screenshot shows no dates and one window only. A CTR of 0.1% at an average position of 56.8 means most impressions came from low positions.
None of the full case results came from more content or more links first. Each one started by removing something that was getting in the way: duplicate addresses, competing pages, inconsistent listings, thin archives, a menu locked in a PDF, or pages that needed scripts to show their text.
That is a pattern I see across most sites, large or small. Google is often already looking at your site and struggling with what it finds. Clearing that up tends to come before anything new is worth building. The guides in my problems section follow the same order, starting with why a store does not rank. The same order applies to where website visits fail to become leads.
Some work I do has no published case yet, and this page says so instead of filling the gap.
This is for you if
Skip this if
If your profile has been suspended, start with the appeal guide for a suspended profile, which explains the process without claiming a result.
Most published cases are for clients outside Pakistan, and the restaurant's location is withheld. They show methods that apply anywhere, but the markets and competitors are different from a business in Lahore or Karachi. When a Pakistani client agrees to publish, their case will be added here with the same rules.
These questions work on my cases and on anyone else's.
What exactly was measured? Clicks, impressions, rankings, and leads are different things. A case about clicks says nothing about sales unless it shows sales.
What was the comparison? Two periods side by side tell you more than one period alone. Seasonal businesses can grow in one quarter for reasons that have nothing to do with SEO.
What was done, and in what order? A list of every SEO task ever invented is not a method. A short list of specific fixes, in order, is.
What are the limits? If a case has no limits, the writer either did not think about them or chose not to tell you. Every real project has limits: missing dates, a different market, several changes made at once, or a measure that stops short of sales.
Could the same thing be true of my site? A case is useful when your site has the same problem. The checks on each case page help you decide.
If your site shows one of the problems in these cases, the method is likely to apply, even though the numbers will differ. A local shop in Lahore with mismatched listings has the same kind of problem the electrician had. A Karachi store with filter pages open to Google has the same kind of problem the international store had.
The fastest way to know is to check. My free website scan looks at the basics in a few minutes. If you would rather talk, a paid hour to review your data is enough to see which case your site resembles. For the full picture, a forensic audit of your site lists every fix in order. Ongoing work is priced on the rupee price list for monthly work.
You can also see how I approach technical problems on larger sites. For a shop or a service area, read about local search for businesses with a physical presence, where the Google Business Profile does most of the work.
Yes. Each one rests on a screenshot from Google Search Console or a map rank tracker, taken from a real client project. Client names are withheld unless the client agrees otherwise.
Most clients do not want competitors to see their search data. I describe each client by industry and market where I can, and show the screenshot with the site address hidden. Clients agree to this in their contract.
I cannot promise any result, and nobody honest can. What I can do is check whether your site has the same kind of problem, and tell you what fixing it would involve.
I only publish a result when I have a screenshot that supports it and the client is comfortable with it being shared. A short list of real cases is more useful than a long list you cannot check.
None that names Pakistan yet. Most cases here are for clients abroad, and the restaurant's location is withheld. Pakistani cases will be added when clients agree and the evidence meets the same rules.
Pick the case that sounds most like your site, read its checks, and run them. If you see the same problem, send me your site and what you found. I reply within 24 hours.
Start with the free automated scan of one page. For a full review by me, send a message from the contact page.
Get my free scan